Abstract: Hyperspectral reflectance data were recorded at main growth stages of cotton in a field experiment including two cultivars with four planting patterns in two years study. By using spectral derivative analysis, red-edge parameters such as area of red edge (SDr), the slope of red edge value (Dr) and corresponding red edge position（λr）were obtained at bands of 680 nm to 750 nm, then the correlation between area of red edge, the slope of red edge and chlorophyll density (CH.D), leaf area index (LAI) were analyzed respectively. The results showed that they were all significant positive correlation at 1% lever; moreover, there was the highest correlation（RCH.D=0.8787**，n=137）between area of red edge and chlorophyll density. Using area of red edge to estimate chlorophyll density, leaf area index, the former had a higher precision than the latter did. Utilizing established model equation between area of red edge and chlorophyll density in cotton varieties of Xin Luzao No.19, Xin Luzao No.13 to derive their chlorophyll density, precision of estimated chlorophyll density of cotton two varieties were 87.4% and 83.3% respectively. The research shows that it is a simple, effective and non-destructive method to use red-edge parameters obtaining the canopy information of CH.D and LAI.